Origin of Life — Checkpoint: The Chromosome
Date: 18/11/2026 08:55 AEDT Follows: origin_of_life_sequence_checkpoint_2026-11-14.md Scope: where, when and how the chromosome formed; what has none; simplest dividing cells and chromosomes; Andrew's plasmid-transfer and "agenetic first genome" positions.
Labels used: [Settled] = mainstream consensus · [Contested] = live scientific debate · [Andrew] = Andrew's own position · [Refined] = a reworded version Claude suggested in this session (not yet adopted)
1. The chromosome formed in three stages, not once
| Stage |
When |
What happened |
Status |
| 1. Genes linked into one strand |
Pre-cell, ~4 billion years ago (date unknown) |
Separate short replicators (probably RNA) joined end to end so they got inherited together. Maynard Smith & Szathmáry (1995) list "origin of chromosomes" as a major evolutionary transition |
[Settled] that it happened; [Contested] when and how |
| 2. DNA takes over → circular chromosome |
Around LUCA (last universal common ancestor), ~4.2 to 3.5 billion years ago |
One circular DNA chromosome with a single copy-start point (most bacteria and archaea; many archaea have several). Bacteria have no histones; many archaea do |
[Contested]: did LUCA store its genes in DNA? Bacterial and archaeal DNA-copying enzymes are largely unrelated, which suggests DNA copying was invented twice. Forterre: viruses invented DNA. Moody et al. 2024: LUCA at ~4.2 billion years ago with a DNA genome (disputed) |
| 3. Eukaryotic chromosome |
~2 to 1.6 billion years ago |
Many linear pieces with end caps (telomeres), histone spools and several copy-start points; arrived with complex cells (Asgard archaeon plus the bacterium that became the mitochondrion) |
[Settled]: histones came from archaea. [Contested]: why linear (telomerase is related to retroelement enzymes; Martin & Koonin 2006 link it to "jumping" introns from the mitochondrion; others tie it to meiosis); how the nucleus formed (inside-out, outside-in, or viral eukaryogenesis from Bell, Livingstone and Takemura) |
2. What has no chromosome
- No dividing cell lineage lacks one. [Settled]
- No chromosome at all: prions (protein only, no nucleic acid), viroids (one bare circle of RNA, 250–400 letters, no protein, no coat), viruses (a genome, but not usually called a chromosome).
- Cells in a body that lose theirs: mature red blood cells, platelets, lens fibre cells and outer skin cells. None of them can divide (Andrew's point). They're dead ends, replaced from stem cells that keep their chromosomes. Each gives up dividing in exchange for a job: room for haemoglobin, transparency, a protective layer.
- Dinoflagellates: have chromosomes but mostly lack histones; they pack DNA with proteins taken from a virus (DVNPs, dinoflagellate/viral nucleoproteins; Gornik et al. 2012). A viral gene replaced core chromosome machinery.
- Naming issue: older textbooks said bacteria have "no true chromosome" (they called it a "nucleoid"). Current usage counts it as a chromosome.
3. The simplest dividing "cells"
| Level |
Example |
How it divides |
| Fewest genes |
JCVI-syn3.0 (2016, 473 genes) → syn3A (2021) |
syn3.0 divided irregularly; adding back a few genes, including ftsZ (the pinching ring), restored normal division. Natural relative: Mycoplasma genitalium (~525 genes) |
| Simplest mechanism |
L-form bacteria (wall-less) |
No division ring needed: they make membrane faster than volume grows, and the surplus membrane bulges and pinches off (Errington lab, 2009–2013) |
| Before life |
Fatty-acid bubbles (Zhu & Szostak 2009) |
Grow into threads and break into daughters under gentle stirring; no genes, no proteins. Matches Andrew's earlier "gentle motion" and "lipids as enablers" positions |
The gap: bubbles show that division can come before any genetic control of it. What they don't show is how a chromosome gets shared out evenly between two daughters.
4. What the simplest chromosome looks like (syn3A / Mycoplasma)
- One closed loop of double-stranded DNA, ~543,000 base pairs; no ends, no telomeres.
- An origin, where copying starts (marked by short repeats that the starter protein DnaA grabs), and a terminus directly opposite. Copying runs both ways round the loop.
- ~493 genes packed almost end to end; most face the same way the copying runs.
- No histones: coiled tight by gyrase and loosely held by a small HU-type packing protein.
- Stretched out, ~185 micrometres long, packed into a cell ~0.4 micrometres across.
- Mycoplasma reads the code UGA as tryptophan; almost every other organism reads it as "stop".
- Smallest linear chromosomes: the nucleomorph of Bigelowiella natans (the shrunken leftover nucleus of a swallowed alga): 3 chromosomes, ~373,000 base pairs in total, with telomeres.
- Fewest chromosomes in an animal: males of the jack jumper ant (Myrmecia pilosula) have one.
5. Corrections settled this session
- A chromosome isn't a protein chain. It is two DNA strands. Each link is a nucleotide (sugar + phosphate + base), and phosphate bridges join the sugars into a sugar-phosphate backbone; there are no peptide bonds. The strands pair (A with T, G with C), and that pairing is what allows copying. Protein chains have no equivalent pairing, so they can't template a copy of their own sequence. Scale: smallest chromosome ~543,000 base pairs vs the largest protein (titin) ~34,000 amino acids.
- Histones aren't the same thing as a chromosome. A histone is a reusable packing protein. Eight form a spool (nucleosome) with ~147 base pairs wrapped round it; H1 clips the DNA where it enters and leaves. Human chromosome 1 carries ~1.2 million spools (~10 million histones); a full human cell has ~31 million spools. Bacteria use no histones; sperm swap most of theirs for protamines; archaea link theirs into a continuous ramp (hypernucleosome), probably closer to the original form.
6. [Andrew] Plasmids exchanged a whole chromosome, and this was the precursor to division
Read as: what plasmids moved was a whole organised unit with its own start point, not loose DNA fragments.
Evidence for it:
- Plasmids really do carry whole chromosomes. When the F plasmid inserts into the E. coli chromosome (an "Hfr" strain), mating drags the chromosome across to the partner, the whole lot in ~100 minutes if left uninterrupted. Jacob & Wollman used this to map E. coli by timed interruption in the 1950s.
- The transfer pump and the division pump are related proteins. FtsK pumps chromosomes into the right daughter at division, and SpoIIIE pumps a whole chromosome into a spore. Both are in the same family as the pump plasmids use to push DNA into a mating partner (VirD4/TraG). Which came first is unknown.
- Plasmids had segregation kits, and chromosomes adopted them. ParM builds an actin-like rod that pushes plasmid copies apart (a miniature spindle). Chromosomes use the same type of kit (ParABS). Vibrio cholerae's second chromosome is a former megaplasmid. Plasmids turning into chromosomes is documented as "chromids".
[Refined]
- Before cells, call them free-moving replicators, not plasmids: "plasmid" only means something relative to a host and its main genome.
- The idea fits best as the precursor to segregation (each daughter getting a copy), bolted onto membranes that were already splitting by physics. Transfer is already copy-then-deliver, with the donor keeping its copy. This fills the gap in section 3.
- In every known case the transferred unit is DNA.
Test: in family-tree comparisons, the conjugation pumps (VirD4/TraG) and plasmid segregation kits should sit deeper (older) than FtsK and the chromosomal kits. This hasn't been settled.
7. [Andrew] The first genome was the chromosome, and the long DNA molecule was agenetic
Word histories:
- Chromosome: Waldeyer, 1888, from Greek chrōma (colour) + sōma (body): "coloured body", because they took up stains. Named before anyone knew they carried heredity (Sutton & Boveri, ~1902–03).
- Chromatin: Flemming, 1879–82, "colourable stuff".
- Gene: Johannsen, 1909, from Greek genos (birth, kind, origin).
- Genome: Winkler, 1920, usually given as Gen + Chromosom. The word "genome" is literally built from "chromosome". Limit: the words record the order of discovery, not the order of evolution, so they illustrate the idea but aren't evidence for it.
[Refined] "Agenetic" needs splitting into two meanings:
- Non-coding (specifies no proteins) is defensible. Ribosomal RNA, transfer RNA and viroids work without coding. DNA lacks RNA's extra oxygen atom, so it's a poor catalyst and suits a quiet structural or storage role.
- Not inherited only holds if the molecule wasn't copied by base-pairing. If it was, its sequence passed down regardless, which makes it genetic in the minimal sense, like viroids.
- Workable statement: the chromosome was first a non-coding structure, and later its sequence was put to work as instructions. The biology term for that is exaptation (a part built for one job taken over for another; feathers for insulation, then flight). On this reading, "the first genome was the chromosome" means a container that turned into the contents' record.
8. Open threads
- What the non-coding chromosome was for before it carried instructions (structure? a scaffold for the amino acid replicator?).
- How an even split of the chromosome between daughters arose: transfer machinery repurposed, or something else.
- Which is older in the family trees: transfer pumps or division pumps.